Re: Ganda / Luganda (forms)

Don Osborn wrote:

> A couple of questions to keep in sight when making the change. First,
> along the lines of Caoimhín's question: What is  precedent wrt
> languages that use prefixes, and the fashion to use endonyms in
> English?

As I tried to explain in the registration form, and in my reply to
Caoimhín, this request has nothing to do with adding endonyms per se,
which would be a violation of RFC 5646 sections 3.1.5 and 3.9.

This is about adding an additional name which is, apparently,
overwhelmingly used to refer to this language in an English-language
setting, i.e. not just an endonym. I don't know if that's some sort of
fashion. People don't normally use the words "français" or "deutsch" or
"русский" when referring to those languages in other languages
AFAIK.

This was precipitated specifically by a CLDR request, which proposed to
override in its English data what the Registry has (making it the sole
name, not just an additional name as I have proposed here) as confirmed
by "several native speakers."

http://unicode.org/cldr/trac/ticket/11004

This is NOT intended to set a precedent for going against the RFC and
adding endonyms wherever we find them.

> It's really not our place IMO to change the English references to
> endonyms. But if, on a case by case basis, the latter start to become
> more common, what's the threshold?

I would say: As we continue to get information that people are using
name B instead of name A to refer to language X, not just when speaking
language X but when speaking other languages, we should consider adding
name B to the Registry.

If it became not just "fashionable," but widespread, for people to start
using the word "français" to refer to the French language, even when
speaking English and other languages, then I would want to consider
adding that as an additional Description field.

Remember that this is not a request to "change" the *primary*
Description field, which came from ISO 639, but to add an *additional*
Description field.

> (The odd result, however, is that you inevitably end up with
> redundancies like "Luganda language" or "ChiBemba language" etc. - tho
> I realize that's an issue beyond IETF's or CLDR's concerns.)

Descriptions in the Registry don't refer to "X language" anyway, except
in the construct "X Sign Language" and for the collections inherited
from ISO 639-5, such as "Bihari languages." We can't otherwise control
how these labels are used grammatically; as you said, even CLDR has
limits in that regard.

> Second is how this works across languages. For example, in Swahili,
> Luganda is Kiganda. Otoh, Chinese has 卢干达语 (lú gān/gàn dá yŭ)
> - Luganda language (effectively Ganda language language) - not clear
> whether this is from the endonym, or from English common usage
> (Chinese in some number of cases adapts English or French versions of
> names of African languages & ethnicities).

If there are any concerns that this request opens a slippery slope to
adding the name of the Ganda/Luganda language in Swahili, or in Chinese,
or that it violates sections 3.1.5 or 3.9, I will be happy to withdraw
the request.


--
Doug Ewell | Thornton, CO, US | ewellic.org


_______________________________________________
Ietf-languages mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ietf-languages